[jeffw@sc47]

This relates to using water, air, or fuel for a fuel tank leak test and some of the points of view I've come across in VAF lately. I had seen a post that a fuel molecule was much smaller than a water molecule and therefore if water did not find a leak, when later filled with fuel you may end up finding one. So I poked around a little.

I'm not a chemist but after spending a while getting 'google' educated on the size or volume of molecules, I am starting this conversation. One page I found was: http://www.chemteam.info/Chem-History/Loschmidt-1865.html

I also talked to two chemical engineers I know and got their opinions/input. What I understood from them, was that a complex carbon molecule such as gasoline was larger than both an air molecule or a water molecule.

Below, a portion of the table in web page address (above) regarding the specific volumes of molecules:

Observed Volume of
Molecules at 0 Celcius

Water H2O 18.0
Octane C8H18 159.7

Showing that a water molecule is significantly smaller in volume than fuel type molecules. In this web page there was discussion about how the interstitial spaces between atoms varies, and viscosity was also mentioned. So, it is apparent that there are more than a few factors that might be considered that may affect how different fuel tank testing materials work.

The diameter of an air molecule: s=8 x 0.000866 x 0.000140 = 0.000000969 mm or in round numbers one millionth of a millimeter. There was no comparison that I could see that provided the diameter in mm for the fuel or water molecules.

I'm sure that there are VAF members who are chemists and can weigh in on this much better than I can. And, who may provide opinions on the appropriateness of varying testing materials to fill tanks with and their ability to find their way through small leak points in a fuel tank assembly.

 

[JonJay]

Other attributes, like surface tension, may be more important than molecule size in this particular application. Water has a very high surface tension and can easily bridge small voids. Fuel has a very low surface tension comparatively.
The size of the holes we are trying to find are not at the molecular level.
Water would be a poor choice of material for testing a tank for this reason compared to alcohol or other fuels.

 

[jeffw@sc47]

Finished, sealed, tested

Finished both tanks about two weeks ago. After water testing and not finding any leaks and installing the rear baffles. After about a week I did the balloon test on both tanks and found a number of leaks at fittings, etc. The leaks were all at the balloon attach, finger strainer fitting, and fuel cap.

Had to tighten one tank's finger strainer fitting about 3/4 turn. After trying to get balloons tightly attached on the vent fittings by wrapping two rubber bands around the connection - there was always a small leak under the balloon at the rubber band wrapping. I finally used a short piece of rubber tubing at the vent fitting with a hose clamp at the fitting end and the balloon at the other end - used a strip piece of inner-tube wrapped around the balloon end to help seal tightly and not damage the balloon material.

The fuel cap on both tanks had leaks; both needed to have their tension nuts tightened a little. I also spread some grease around the o-ring. Both tanks will hold a balloon inflated for days - so I leave them inflated to make me happy. Pretty confident the tanks are tight.

The balloon adjust to the day/night temperature swings and every few days a balloon will pop. They are higher quality balloons - I get them from a 'five and dime' store that does lots of helium filled balloons. So I replace it and it stays inflated until another 'pop'.